
import numpy as np
import matplotlib.pyplot as plt
import matplotlib
import seaborn as sns
import pandas as pd
pd.set_option('display.max_rows', 500)
pd.set_option('display.max_columns', 500)
pd.set_option('display.width', 1000)
plt.style.use('ggplot')


df = pd.read_csv("../data/Spotify_Data.csv")


type(df)

pandas.core.frame.DataFrame


df.head(10)


df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 953 entries, 0 to 952
Data columns (total 25 columns):
 #   Column                Non-Null Count  Dtype 
---  ------                --------------  ----- 
 0   track_name            953 non-null    object
 1   artist(s)_name        953 non-null    object
 2   artist_count          953 non-null    int64 
 3   released_year         953 non-null    int64 
 4   released_month        953 non-null    int64 
 5   released_day          953 non-null    int64 
 6   in_spotify_playlists  953 non-null    int64 
 7   in_spotify_charts     953 non-null    int64 
 8   streams               953 non-null    object
 9   in_apple_playlists    953 non-null    int64 
 10  in_apple_charts       953 non-null    int64 
 11  in_deezer_playlists   953 non-null    object
 12  in_deezer_charts      953 non-null    int64 
 13  in_shazam_charts      903 non-null    object
 14  bpm                   953 non-null    int64 
 15  key                   858 non-null    object
 16  mode                  953 non-null    object
 17  danceability_%        953 non-null    int64 
 18  valence_%             953 non-null    int64 
 19  energy_%              953 non-null    int64 
 20  acousticness_%        953 non-null    int64 
 21  instrumentalness_%    953 non-null    int64 
 22  liveness_%            953 non-null    int64 
 23  speechiness_%         953 non-null    int64 
 24  cover_url             953 non-null    object
dtypes: int64(17), object(8)
memory usage: 186.3+ KB


print(df.duplicated().sum()) #Check for duplicates

0


""" Summarizing data """

def column_summary(df):
    summary_data = []
    for col_name in df.columns:
        col_dtype = df[col_name].dtype
        num_of_nulls = df[col_name].isnull().sum()
        num_of_non_nulls = df[col_name].notnull().sum()
        num_of_distinct_values = df[col_name].nunique()

        if num_of_distinct_values <= 10:
            distinct_values_counts = df[col_name].value_counts().to_dict()
        else:
            top_10_values_counts = df[col_name].value_counts().head(10).to_dict()
            distinct_values_counts = {k: v for k, v in sorted(top_10_values_counts.items(), key= lambda item: item[1], reverse=True )}
        summary_data.append({
            'col_name': col_name,
            'col_dtype': col_dtype,
            'num_of_nulls': num_of_nulls,
            'num_of_non_nulls': num_of_non_nulls,
            'num_of_distinct_values': num_of_distinct_values,
            'distinct_values_counts': distinct_values_counts
        })
    summary_df = pd.DataFrame(summary_data)
    return summary_df

summary_df = column_summary(df)
display(summary_df)


""" Transforming data types & eliminating useless features"""

df["streams"] = pd.to_numeric(df["streams"],errors="coerce") #There is an entry containing metadata.
df[["key","mode"]] = df[["key","mode"]].astype("category") #Categorical features
df = df.drop(["cover_url","in_deezer_playlists","in_deezer_charts","in_shazam_charts"],axis=1) #Useless features
df = df.dropna()
df["streams"] = df["streams"].astype("int64")


summary_df = column_summary(df)
display(summary_df)


df.describe() #Basic statistics


numerical_columns = df.select_dtypes("int64").columns
num_columns = len(numerical_columns)
num_rows = (num_columns + 2) // 3  # Compute number of rows needed for 3 columns per row

# Create figure and axes
fig, axes = plt.subplots(num_rows, 3, figsize=(18, 5 * num_rows))

# Flatten axes array for easy indexing
axes = axes.flatten()

for i, column in enumerate(numerical_columns):
    ax = axes[i]
    
    if df[column].nunique() > 10:  # Continuous variable (>10 unique values)
        sns.histplot(df[column], color="blue", kde=True, ax=ax) #kde=True -> plot an approximate probability density function
        ax.set_title(f"Histogram of {column}")
        ax.set_xlabel(column)
        ax.set_ylabel("Frequency")
    else:  # Discrete variable (<10 unique values)
        sns.countplot(x=column, data=df, color="blue", ax=ax)
        ax.set_title(f"Count of {column}")
        ax.set_xlabel(column)
        ax.set_ylabel("Count")

        # Add value labels on bars
        for p in ax.patches:
            ax.annotate(format(p.get_height(), ".0f"),
                        (p.get_x() + p.get_width() / 2., p.get_height()),
                        ha="center", va="bottom",
                        xytext=(0, 5),
                        textcoords="offset points")

# Remove any unused subplots
for j in range(i + 1, len(axes)):
    fig.delaxes(axes[j])

plt.tight_layout()
plt.show()


correlation_matrix = df[numerical_columns].corr()
print(correlation_matrix)

                      artist_count  released_year  released_month  released_day  in_spotify_playlists  in_spotify_charts   streams  in_apple_playlists  in_apple_charts       bpm  danceability_%  valence_%  energy_%  acousticness_%  instrumentalness_%  liveness_%  speechiness_%
artist_count              1.000000       0.076616        0.032990     -0.011917             -0.100919          -0.020673 -0.131273           -0.055404        -0.095824 -0.063419        0.207636   0.117233  0.135091       -0.101232           -0.055208    0.032083       0.131129
released_year             0.076616       1.000000        0.083459      0.171435             -0.387940           0.068413 -0.222450           -0.196616        -0.029040 -0.010362        0.200100  -0.055269  0.097239       -0.150987           -0.020374   -0.011370       0.134253
released_month            0.032990       0.083459        1.000000      0.068443             -0.109876          -0.037911 -0.026944           -0.010285        -0.017907 -0.043506       -0.063124  -0.114357 -0.077919        0.043163            0.034562   -0.003034       0.045511
released_day             -0.011917       0.171435        0.068443      1.000000             -0.088851           0.036835 -0.000371            0.009003         0.004471 -0.021101        0.064164   0.056769  0.053526       -0.009193            0.023780   -0.021384      -0.016784
in_spotify_playlists     -0.100919      -0.387940       -0.109876     -0.088851              1.000000           0.173381  0.787825            0.708076         0.259238 -0.018566       -0.103279  -0.021648  0.039666       -0.056214           -0.031697   -0.049014      -0.090344
in_spotify_charts        -0.020673       0.068413       -0.037911      0.036835              0.173381           1.000000  0.252665            0.234889         0.551774  0.039211        0.030249   0.036794  0.105216       -0.064086           -0.007873   -0.029890      -0.098284
streams                  -0.131273      -0.222450       -0.026944     -0.000371              0.787825           0.252665  1.000000            0.774445         0.313651 -0.002163       -0.100702  -0.042787 -0.029714        0.010798           -0.043068   -0.054275      -0.112896
in_apple_playlists       -0.055404      -0.196616       -0.010285      0.009003              0.708076           0.234889  0.774445            1.000000         0.408485  0.030178       -0.025617   0.052390  0.045562       -0.050028           -0.061521   -0.063207      -0.108237
in_apple_charts          -0.095824      -0.029040       -0.017907      0.004471              0.259238           0.551774  0.313651            0.408485         1.000000  0.034633       -0.028701   0.046622  0.124038       -0.088594           -0.007183   -0.014382      -0.150397
bpm                      -0.063419      -0.010362       -0.043506     -0.021101             -0.018566           0.039211 -0.002163            0.030178         0.034633  1.000000       -0.161907   0.027436  0.016068       -0.012766           -0.005266   -0.003106       0.046373
danceability_%            0.207636       0.200100       -0.063124      0.064164             -0.103279           0.030249 -0.100702           -0.025617        -0.028701 -0.161907        1.000000   0.397807  0.166331       -0.240092           -0.099225   -0.093882       0.196831
valence_%                 0.117233      -0.055269       -0.114357      0.056769             -0.021648           0.036794 -0.042787            0.052390         0.046622  0.027436        0.397807   1.000000  0.346820       -0.066017           -0.144168    0.010830       0.040850
energy_%                  0.135091       0.097239       -0.077919      0.053526              0.039666           0.105216 -0.029714            0.045562         0.124038  0.016068        0.166331   0.346820  1.000000       -0.561660           -0.050323    0.103592      -0.025340
acousticness_%           -0.101232      -0.150987        0.043163     -0.009193             -0.056214          -0.064086  0.010798           -0.050028        -0.088594 -0.012766       -0.240092  -0.066017 -0.561660        1.000000            0.043322   -0.043199      -0.019821
instrumentalness_%       -0.055208      -0.020374        0.034562      0.023780             -0.031697          -0.007873 -0.043068           -0.061521        -0.007183 -0.005266       -0.099225  -0.144168 -0.050323        0.043322            1.000000   -0.044868      -0.088203
liveness_%                0.032083      -0.011370       -0.003034     -0.021384             -0.049014          -0.029890 -0.054275           -0.063207        -0.014382 -0.003106       -0.093882   0.010830  0.103592       -0.043199           -0.044868    1.000000      -0.015711
speechiness_%             0.131129       0.134253        0.045511     -0.016784             -0.090344          -0.098284 -0.112896           -0.108237        -0.150397  0.046373        0.196831   0.040850 -0.025340       -0.019821           -0.088203   -0.015711       1.000000


#A more visual representation with a heat map

plt.figure(figsize=(16,12))
sns.heatmap(correlation_matrix,
            annot = True, #Annotate numerical value in each cell
            cmap="coolwarm", #chosen color gradient (blue=negative, white=zero, red=positive)
            fmt=".2f") #formats numerical values to 2 decimal places in the annotations

plt.title("Correlation Heatmap")
plt.tight_layout() #Adjust plot layout to avoid title/label overlap
plt.show()

# Find the max correlation
upper_triangular = correlation_matrix.where(np.triu(np.ones(correlation_matrix.shape), k=1).astype(bool)) #This is a very compact way of selecting only the upper triangular part of the correlation matrix, which is the only thing that matters.
                                                                                                          # Remember: correlation is symmetrical, and the self-correlation is always 1.
                                                                                                          # What we are doing is creating an upper triangular matrix of "True" values (excluding the diagonal)
                                                                                                          # Using the .where(...) we are effectively selecting these same positions from the correlation matrix.
max_correlation = upper_triangular.max().max() #We have to use .max() twice because using it once returns the max value for each column of the correlation matrix. We want a single value.
print(f"Maximum pairwise correlation: {max_correlation:.2f}")

Maximum pairwise correlation: 0.79


#Exploring pairs of musical features vs success (streams)
df["streams_bin"] = pd.qcut(df["streams"], q=4, labels=["Low", "Mid", "High", "Top"])
palette = {
    "Low": "#6b1d1d",   # blue
    "Mid": "#a83131",   # green
    "High": "#e77110",  # orange
    "Top": "#ffae00",   # red
}

sns.pairplot(data=df,
             vars=["bpm","danceability_%","valence_%","energy_%","acousticness_%","instrumentalness_%","liveness_%","speechiness_%"],
             hue="streams_bin",
             palette=palette)

plt.show()


"Analyze categorical feature: key"

sns.countplot(x="key",data=df,hue="key")
plt.xlabel("Keys",fontsize=16)
plt.ylabel("Count",fontsize=16)
plt.title("Number of tracks for each key",fontsize=18)
plt.show()


x_vals = ["streams","in_spotify_charts","in_spotify_playlists"]
titles = ["Streams","Popularity","Engagement"]

for i,x in enumerate(x_vals):
    sns.catplot(y="key",
                x=x,
                data=df,
                aspect=2,
                hue="key",
                legend=False)
    plt.xlabel(titles[i], fontsize=16)
    plt.ylabel("Keys",fontsize=16)
plt.show()


fig, axes = plt.subplots(1,3,figsize=(16,6))

axes = axes.flatten()

sns.boxplot(y="key",
            x="streams",
            data=df,
            hue="key",
            fliersize=2,
            ax=axes[0])
axes[0].set_xlabel("Streams", fontsize=12)
axes[0].set_ylabel("Keys",fontsize=12)

sns.boxplot(y="key",
            x="in_spotify_charts",
            data=df,
            hue="key",
            fliersize=2,
            ax=axes[1])
axes[1].set_xlabel("Popularity", fontsize=12)

sns.boxplot(y="key",
            x="in_spotify_playlists",
            data=df,
            hue="key",
            fliersize=2,
            ax=axes[2])
axes[2].set_xlabel("Engagement", fontsize=12)

plt.show()


"Analyze categorical feature: mode"

sns.countplot(x="mode",data=df,hue="mode")
plt.xlabel("Mode",fontsize=16)
plt.ylabel("Count",fontsize=16)
plt.title("Number of tracks for each mode",fontsize=18)
plt.show()


palette = sns.color_palette("Set2", len(df["mode"].unique()))

sns.catplot(y="mode",
            x="streams",
            data=df,
            aspect=2,
            palette=palette)
plt.xlabel("Streams", fontsize=20)
plt.ylabel("Mode",fontsize=20)
plt.show()

C:\Users\pablo\AppData\Local\Temp\ipykernel_5460\3483730756.py:3: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect.

  sns.catplot(y="mode",


fig, axes = plt.subplots(1,3,figsize=(16,4))

axes = axes.flatten()

sns.boxplot(y="mode",
            x="streams",
            data=df,
            hue="mode",
            fliersize=2,
            ax=axes[0])
axes[0].set_xlabel("Streams", fontsize=12)
axes[0].set_ylabel("Mode",fontsize=12)

sns.boxplot(y="mode",
            x="in_spotify_charts",
            data=df,
            hue="mode",
            fliersize=2,
            ax=axes[1])
axes[1].set_xlabel("Popularity", fontsize=12)

sns.boxplot(y="mode",
            x="in_spotify_playlists",
            data=df,
            hue="mode",
            fliersize=2,
            ax=axes[2])
axes[2].set_xlabel("Engagement", fontsize=12)

plt.show()


tops = [5,10,25,50,75,100]
labels = ["Top 5", "Top 10", "Top 25", "Top 50", "Top 75", "Top 100"]

def analyze_top(metric, feature, numerical=True):
    """ Function for visualizing features grouped by success metrics.
        Args:
            metric (str): success metric.
            feature (str): feature to visualize.
            numerical (bool): feature is numerical (True) or categorical (False)
    """
    if numerical: #Numerical feature: plot data points + mean tendency
        plot_data = []

        for top, label in zip(tops, labels):
            data = df.nlargest(top, metric)

            temp_df = pd.DataFrame({
                "top": [label] * len(data),
                "values": data[feature]
            })
            plot_data.append(temp_df)

        point_df = pd.concat(plot_data, ignore_index=True) #Concatenate df for different "top" values

        plt.figure(figsize=(8, 6))
        sns.stripplot( #Plot data points for each subgroup
            data=point_df,
            x="top",
            y="values",
            color="blue",
            alpha=.25,
            legend=False
        )
        sns.pointplot( #Compute mean for each subgroup + tendency
            data=point_df,
            x="top",
            y="values",
            marker="D",
            markersize=5, 
            linestyle="--",
            errorbar=None,
            color="blue"            
        )

        plt.title(f"Mean of {feature} across Top Tracks by {metric}")
        plt.ylabel(f"{feature}")
        plt.xticks(rotation=45)
        plt.tight_layout()
        plt.show()
        

    else: #Categorical feature
        plot_data = []

        for top in tops:
            data = df.nlargest(top, metric)
            top_5 = data[feature].value_counts().head(5) #Extract top 5 most ocurring categories
            top_5_relative = {k: v / top * 100 for k, v in top_5.items()} #Compute relative frequency
            plot_data.append(top_5_relative) #Plot data is now a list of dictionaries

        # Collect all top categories
        all_categories = set() #Creating a set ensures that categories aren't repeated
        for d in plot_data:
            all_categories.update(d.keys())
        all_categories = list(all_categories)

        # Create frequency matrix
        freq_matrix = []
        for d in plot_data:
            row = [d.get(cat, 0) for cat in all_categories]
            freq_matrix.append(row) #Each row of freq_matrix contains the relative frequencies for a fixed "top" value in (5,10,25,etc.)
                                    #each column represents the relative frequencies of a fixed "category" across all "top" values

        freq_df = pd.DataFrame(freq_matrix, columns=all_categories, index=labels)

        ax = freq_df.plot(kind="bar", figsize=(12, 6), colormap="Set2")
        plt.title(f"{feature} % across Top Tracks by {metric}")
        plt.ylabel("Relative Frequency (%)")
        plt.xticks(rotation=45)
        plt.legend(title="Category", bbox_to_anchor=(1.05, 1), loc='upper left')
        plt.tight_layout()
        plt.show()


numerical_columns = ["bpm","danceability_%","valence_%","energy_%","acousticness_%","instrumentalness_%","liveness_%","speechiness_%"]
for column in numerical_columns:
    analyze_top("in_spotify_playlists",column,numerical=True)


categorical_columns = ["key","mode"]
for column in categorical_columns:
    analyze_top("in_spotify_playlists",column,numerical=False)


for column in numerical_columns:
    analyze_top("in_spotify_charts",column,numerical=True)
for column in categorical_columns:
    analyze_top("in_spotify_charts",column,numerical=False)


for column in numerical_columns:
    analyze_top("streams",column,numerical=True)
for column in categorical_columns:
    analyze_top("streams",column,numerical=False)


#Get names of individual artists
artists = {name.strip() for names in df["artist(s)_name"] for name in names.split(",") if name.strip()}

#Collect their stats (numerical features)
artists_avg = {}

for artist in artists:
    subset_df = df[df["artist(s)_name"].str.contains(artist, regex=False)] #regex=False avoids special character issues
    number_of_tracks = len(subset_df)
    total_streams = subset_df["streams"].sum()
    stats = subset_df.select_dtypes(include="number").mean()
    stats["number_of_tracks"] = number_of_tracks
    stats["total_streams"] = total_streams

    artists_avg[artist] = stats

artists_avg_df = pd.DataFrame.from_dict(artists_avg,orient="index")

top_artists = artists_avg_df.sort_values("total_streams",ascending=False).reset_index().rename(columns={"index":"artist"}).head(18)


top_artists = top_artists[~top_artists["artist"].isin(["M", "V", "Bad B"])].reset_index(drop=True)
#Mismatchs that we need to remove...
#"M", "V", and "Bad B" are artists, but clearly their stream count is that high because their name is contained in other artist's names


#We only look at musical features and total_streams
top_artists = top_artists.drop(columns=["released_year","released_month","released_day","streams","in_spotify_playlists","in_spotify_charts","in_apple_playlists","in_apple_charts","number_of_tracks"],axis=1)


plt.figure(figsize=(10,6))
sns.barplot(data=top_artists,x="artist",y="total_streams",hue="artist")
plt.xticks(fontsize=8, rotation=45)
plt.ylabel("Total streams")
plt.title("Top 15 artists by total streams")

plt.tight_layout()
plt.show()


plt.figure(figsize=(18,20))
top_artists.hist(color="blue",bins=10)
plt.tight_layout()
plt.show()

<Figure size 1800x2000 with 0 Axes>

	artist_count	released_year	released_month	released_day	in_spotify_playlists	in_spotify_charts	streams	in_apple_playlists	in_apple_charts	bpm	danceability_%	valence_%	energy_%	acousticness_%	instrumentalness_%	liveness_%	speechiness_%
count	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000	8.570000e+02	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000	857.000000
mean	1.551925	2018.297550	6.031505	13.739790	5204.997666	11.959160	5.133554e+08	67.383897	51.044341	122.842474	67.273046	51.219370	64.339557	26.644107	1.611435	18.175029	10.396733
std	0.864634	10.991135	3.567138	9.287652	7944.032885	19.194211	5.714855e+08	86.562405	50.407409	28.196567	14.653159	23.594579	16.056348	25.688111	8.570734	13.569471	10.090890
min	1.000000	1930.000000	1.000000	1.000000	31.000000	0.000000	2.762000e+03	0.000000	0.000000	65.000000	23.000000	4.000000	14.000000	0.000000	0.000000	3.000000	2.000000
25%	1.000000	2020.000000	3.000000	5.000000	859.000000	0.000000	1.391938e+08	13.000000	7.000000	100.000000	57.000000	32.000000	53.000000	5.000000	0.000000	10.000000	4.000000
50%	1.000000	2022.000000	5.000000	13.000000	2226.000000	3.000000	2.849083e+08	34.000000	38.000000	121.000000	70.000000	51.000000	66.000000	17.000000	0.000000	12.000000	6.000000
75%	2.000000	2022.000000	9.000000	22.000000	5542.000000	16.000000	6.740727e+08	85.000000	85.000000	142.000000	78.000000	70.000000	76.000000	42.000000	0.000000	24.000000	12.000000
max	8.000000	2023.000000	12.000000	31.000000	52898.000000	147.000000	3.703895e+09	672.000000	275.000000	206.000000	96.000000	97.000000	97.000000	97.000000	91.000000	97.000000	64.000000

Report: Exploratory Data Analysis on "Most streamed tracks in Spotify 2023" dataset.¶

Data collection¶

Data cleaning & pre-processing¶

Comments¶

Comments¶

Univariate analysis¶

Comments¶

Correlation analysis¶

Comments¶

Multivariate analysis¶

Note¶

Comments¶

Comments¶

Top tracks analysis¶

Comments¶

Top artists analysis¶

Comments¶

	track_name	artist(s)_name	artist_count	released_year	released_month	released_day	in_spotify_playlists	in_spotify_charts	streams	in_apple_playlists	in_apple_charts	in_deezer_playlists	in_deezer_charts	in_shazam_charts	bpm	key	mode	danceability_%	valence_%	energy_%	acousticness_%	instrumentalness_%	liveness_%	speechiness_%	cover_url
0	Seven (feat. Latto) (Explicit Ver.)	Latto, Jung Kook	2	2023	7	14	553	147	141381703	43	263	45	10	826	125	B	Major	80	89	83	31	0	8	4	Not Found
1	LALA	Myke Towers	1	2023	3	23	1474	48	133716286	48	126	58	14	382	92	C#	Major	71	61	74	7	0	10	4	https://i.scdn.co/image/ab67616d0000b2730656d5...
2	vampire	Olivia Rodrigo	1	2023	6	30	1397	113	140003974	94	207	91	14	949	138	F	Major	51	32	53	17	0	31	6	https://i.scdn.co/image/ab67616d0000b273e85259...
3	Cruel Summer	Taylor Swift	1	2019	8	23	7858	100	800840817	116	207	125	12	548	170	A	Major	55	58	72	11	0	11	15	https://i.scdn.co/image/ab67616d0000b273e787cf...
4	WHERE SHE GOES	Bad Bunny	1	2023	5	18	3133	50	303236322	84	133	87	15	425	144	A	Minor	65	23	80	14	63	11	6	https://i.scdn.co/image/ab67616d0000b273ab5c9c...
5	Sprinter	Dave, Central Cee	2	2023	6	1	2186	91	183706234	67	213	88	17	946	141	C#	Major	92	66	58	19	0	8	24	https://i.scdn.co/image/ab67616d0000b273e3a09a...
6	Ella Baila Sola	Eslabon Armado, Peso Pluma	2	2023	3	16	3090	50	725980112	34	222	43	13	418	148	F	Minor	67	83	76	48	0	8	3	https://i.scdn.co/image/ab67616d0000b2732071a0...
7	Columbia	Quevedo	1	2023	7	7	714	43	58149378	25	89	30	13	194	100	F	Major	67	26	71	37	0	11	4	https://i.scdn.co/image/ab67616d0000b273a00a81...
8	fukumean	Gunna	1	2023	5	15	1096	83	95217315	60	210	48	11	953	130	C#	Minor	85	22	62	12	0	28	9	https://i.scdn.co/image/ab67616d0000b273017d5e...
9	La Bebe - Remix	Peso Pluma, Yng Lvcas	2	2023	3	17	2953	44	553634067	49	110	66	13	339	170	D	Minor	81	56	48	21	0	8	33	Not Found

	col_name	col_dtype	num_of_nulls	num_of_non_nulls	num_of_distinct_values	distinct_values_counts
0	track_name	object	0	953	943	{'Flowers': 2, 'Daylight': 2, 'Let It Snow! Le...
1	artist(s)_name	object	0	953	645	{'Taylor Swift': 34, 'The Weeknd': 22, 'Bad Bu...
2	artist_count	int64	0	953	8	{1: 587, 2: 254, 3: 85, 4: 15, 5: 5, 6: 3, 8: ...
3	released_year	int64	0	953	50	{2022: 402, 2023: 175, 2021: 119, 2020: 37, 20...
4	released_month	int64	0	953	12	{1: 134, 5: 128, 3: 86, 6: 86, 11: 80, 12: 75,...
5	released_day	int64	0	953	31	{1: 95, 21: 44, 13: 43, 24: 40, 7: 39, 2: 39, ...
6	in_spotify_playlists	int64	0	953	879	{1473: 3, 685: 3, 3006: 3, 892: 3, 896: 3, 115...
7	in_spotify_charts	int64	0	953	82	{0: 405, 4: 48, 2: 42, 6: 36, 3: 18, 8: 17, 5:...
8	streams	object	0	953	949	{'395591396': 2, '156338624': 2, '1223481149':...
9	in_apple_playlists	int64	0	953	234	{0: 23, 8: 22, 16: 20, 20: 20, 10: 20, 4: 20, ...
10	in_apple_charts	int64	0	953	172	{0: 100, 1: 40, 2: 26, 3: 24, 6: 16, 5: 15, 21...
11	in_deezer_playlists	object	0	953	348	{'0': 24, '15': 23, '5': 20, '13': 20, '12': 1...
12	in_deezer_charts	int64	0	953	34	{0: 558, 1: 137, 2: 48, 3: 31, 6: 18, 4: 18, 5...
13	in_shazam_charts	object	50	903	198	{'0': 344, '1': 73, '2': 35, '3': 21, '4': 19,...
14	bpm	int64	0	953	124	{120: 39, 130: 31, 140: 31, 92: 25, 110: 24, 1...
15	key	object	95	858	11	{'C#': 120, 'G': 96, 'G#': 91, 'F': 89, 'B': 8...
16	mode	object	0	953	2	{'Major': 550, 'Minor': 403}
17	danceability_%	int64	0	953	72	{70: 43, 77: 32, 80: 31, 56: 30, 74: 29, 81: 2...
18	valence_%	int64	0	953	94	{24: 21, 40: 20, 59: 18, 55: 18, 53: 18, 61: 1...
19	energy_%	int64	0	953	80	{74: 29, 62: 28, 76: 27, 66: 25, 73: 23, 68: 2...
20	acousticness_%	int64	0	953	98	{0: 60, 1: 48, 4: 35, 2: 33, 3: 30, 5: 30, 9: ...
21	instrumentalness_%	int64	0	953	39	{0: 866, 1: 21, 2: 7, 4: 5, 3: 4, 5: 4, 18: 3,...
22	liveness_%	int64	0	953	68	{11: 102, 9: 93, 10: 78, 12: 72, 8: 54, 13: 47...
23	speechiness_%	int64	0	953	48	{4: 175, 3: 152, 5: 130, 6: 76, 8: 52, 7: 49, ...
24	cover_url	object	0	953	535	{'Not Found': 225, 'https://i.scdn.co/image/ab...

	col_name	col_dtype	num_of_non_nulls	num_of_distinct_values	distinct_values_counts
0	track_name	object	857	851	{'About Damn Time': 2, 'Die For You': 2, 'Let ...
1	artist(s)_name	object	857	596	{'Taylor Swift': 29, 'The Weeknd': 17, 'SZA': ...
2	artist_count	int64	857	8	{1: 526, 2: 231, 3: 76, 4: 14, 5: 5, 6: 3, 8: ...
3	released_year	int64	857	47	{2022: 362, 2023: 159, 2021: 107, 2019: 33, 20...
4	released_month	int64	857	12	{1: 120, 5: 118, 6: 78, 3: 76, 12: 70, 11: 69,...
5	released_day	int64	857	31	{1: 87, 21: 39, 2: 38, 4: 38, 6: 37, 13: 36, 2...
6	in_spotify_playlists	int64	857	796	{896: 3, 86: 3, 356: 3, 811: 3, 892: 3, 685: 2...
7	in_spotify_charts	int64	857	78	{0: 364, 4: 42, 2: 37, 6: 33, 8: 17, 3: 16, 1:...
8	streams	int64	857	854	{723894473: 2, 1223481149: 2, 156338624: 2, 84...
9	in_apple_playlists	int64	857	224	{8: 21, 0: 20, 20: 19, 10: 19, 4: 19, 5: 18, 7...
10	in_apple_charts	int64	857	168	{0: 93, 1: 35, 3: 23, 2: 22, 6: 15, 5: 14, 7: ...
11	bpm	int64	857	123	{120: 34, 130: 28, 140: 26, 92: 22, 150: 21, 1...
12	key	category	857	11	{'C#': 120, 'G': 96, 'G#': 91, 'F': 89, 'D': 8...
13	mode	category	857	2	{'Major': 474, 'Minor': 383}
14	danceability_%	int64	857	72	{70: 37, 80: 30, 56: 27, 77: 27, 74: 27, 81: 2...
15	valence_%	int64	857	94	{24: 21, 40: 19, 55: 17, 53: 16, 59: 16, 61: 1...
16	energy_%	int64	857	78	{74: 28, 62: 28, 76: 26, 66: 24, 73: 23, 60: 2...
17	acousticness_%	int64	857	95	{0: 52, 1: 48, 4: 33, 2: 30, 3: 29, 9: 28, 5: ...
18	instrumentalness_%	int64	857	36	{0: 780, 1: 19, 2: 5, 3: 4, 5: 4, 4: 4, 63: 3,...
19	liveness_%	int64	857	67	{11: 93, 9: 82, 10: 69, 12: 65, 8: 50, 13: 43,...
20	speechiness_%	int64	857	48	{4: 155, 3: 130, 5: 113, 6: 69, 8: 47, 7: 45, ...